Voltage gated sodium channels (NaV's) are critical determinants of cellular excitability in muscle and nerve (Hille, B, Ion Channels of Excitable Membranes (2001), Sunderland, Mass., Sinauer Associates, Inc.). Four isoforms in particular, NaV1.1, NaV1.2, NaV1.3, and NaV1.6, account for the majority of sodium current in the neurons of the central nervous system. NaV1.3 is primarily expressed embryonically. Beyond the neonatal stage, NaV1.1, NaV1.2, and NaV1.6 are the critical isoforms that regulate neuronal signaling in the brain (Catterall, W. A., Annual Review of Pharmacology and Toxicology (2014), Vol. 54, pp. 317-338).
NaV1.5 is expressed mainly in cardiac myocytes (Raymond, C. K. et al., J. Biol. Chem. (2004), Vol. 279, No. 44, pp. 46234-41), including atria, ventricles, the sino-atrial node, atrio-ventricular node and cardiac Purkinje fibers. Mutations in human NaV1.5 result in multiple arrhythmic syndromes, including, for example, long QT3 (LQT3), Brugada syndrome (BS), an inherited cardiac conduction defect, sudden unexpected nocturnal death syndrome (SUNDS) and sudden infant death syndrome (SIDS) (Liu, H., et al., Am. J. Pharmacogenomics (2003), Vol. 3, No. 3, pp. 173-9). Sodium channel blocker therapy has been used extensively in treating cardiac arrhythmias.
Epilepsy is a condition characterized by excessive synchronous excitability in the brain that arises when the delicate balance of excitatory and inhibitory signals in the brain fall out of equilibrium. This can happen either due to an excess of excitation, or a deficiency of inhibition. Mutations in the genes encoding NaV channels have been linked to both types of disequilibrium.
NaV1.1 has been identified as the primary NaV isoform of inhibitory interneurons (Yu, F. H. et al., Nat. Neurosci. (2006), Vol. 9, pp. 1142-1149). These interneurons synapse on many other neurons, including excitatory glutamatergic neurons. Action potentials in the interneurons induce the release of the neurotransmitter GABA onto other neurons, hyperpolarizing them and thus dampening excitation. This results in a negative feedback that enables controlled signaling and prevents local signals from expanding into waves of excitation that spread across large brain regions. Because of this critical role in inhibitory interneurons, mutations that impair NaV1.1 channel function can lead to a failure of those neurons to activate and release GABA (Ogiwara, I. et al., J. Neurosci. (2007), Vol. 27, pp. 5903-5914; Martin, M. S. et al., J. Biol. Chem. (2010), Vol. 285, pp. 9823-9834; Cheah, C. S. et al., Channels (Austin) (2013), Vol. 7, pp. 468-472; and Dutton, S. B., et al., (2013), Vol. 49, pp. 211-220). The result is a loss in the inhibitory tone of the brain and a failure to contain the excitability of the glutamatergic neurons. This failure of the inhibitory interneurons can result in aberrant wide-scale synchronous firing of neurons across regions of the brain (epilepsy).
Mutations in the gene encoding NaV1.1 (SCN1A) fall into two broad classes, those that cause generalized epilepsy with febrile seizures plus (GEFS+) and those that cause severe myoclonic epilepsy of infancy (SMEI), also known as Dravet Syndrome or early infantile epileptic encephalopathy 6 (EIEE6) (McKusik, V. K. et al., A Epileptic Encephalopathy, Early Infantile 6, EIEE6 (2012), Online Mendelian Inheritance in Man: John Hopkins University). SMEI mutations are heterozygous autosomal dominant mutations and are often caused by a gene deletion or truncation that leads to a channel with little or no function. The mutations arise de novo, or in a few cases have been shown to arise in asymptomatic mosaic parents (Tuncer, F. N. et al., Epilepsy Research (2015), Vol. 113, pp. 5-10). Patients are born phenotypically normal and meet developmental milestones until the onset of seizures, typically between the age of 6 months and 1 year. This time of onset is believed to be a consequence of the normal decrease in the expression of the embryonic isoform NaV1.3 and the coincident rise of NaV1.1. When the NaV1.1 channels fail to reach normal levels, the phenotype is revealed (Cheah, C. S. et al., Channels (Austin) (2013), Vol. 7, pp. 468-472). The initial seizure is often triggered by a febrile episode and can manifest as status epilepticus. Seizures continue and increase in frequency and severity for the first several years of life and can reach frequencies of over 100 episodes per day. Seizures may be triggered by fever or may arise spontaneously without apparent cause. After seizure onset patients begin to miss developmental milestones and significant cognitive and behavioral deficits accrue (Dravet, C. and Oguni, H., Handbook of Clinical Neurology (2013), Vol. 111, pp. 627-633). 80 to 85% of phenotypically diagnosed Dravet syndrome patients are believed to have a responsible mutation in SCN1A, while the other 15-20% of patients have other mutations or are of unknown etiology. There is a high rate of sudden unexplained death in epilepsy (SUDEP) in SMEI patients, with an estimated 37% of patients dying by SUDEP, but the mechanism for this catastrophic outcome remains unclear (Massey, C. A., et al., Nature Reviews Neurology (2014), Vol. 10, pp. 271-282). Clinically useful anti-epileptic drugs that target voltage-gated sodium channels non-selectively, like carbamazepine and phenytoin, are contra-indicated for SMEI patients as they can exacerbate seizures in these patients (Wilmshurst, J. M. et al., Epilepsia (2015), Vol. 56, pp. 1185-1197). This is presumed to be because patients cannot tolerate further reductions in NaV1.1 function.
GEFS+ is often caused by missense SCN1A mutations that induce relatively mild channel dysfunction, consistent with the relatively milder seizure phenotype. A large and growing number of mutations have been identified, and both the severity and the penetrance of the phenotype varies considerably. Many GEFS+ patients outgrow the seizure phenotype, however not all do, and GEFS+ patients with childhood epilepsy are considerably more prone to have epilepsy as adults than are the general population. Mutations that cause deficits in other genes involved with GABA-ergic signaling, like SCN1B that encodes the sodium channel auxiliary subunit and GABRG2 that encodes a subunit of GABAA receptors can also give rise to GEFS+(Helbig, I., Seminars in Neurology (2015) Vol. 35, pp. 288-292).
Transgenic mice have been developed that harbor the same mutations identified in SMEI and GEFS+ patients. In both cases the mice replicate the human phenotype well, though the penetrance of the phenotype can be significantly impacted by the genetic background. Some mouse strains tolerate the mutations relatively well, while in other strains the same mutations can cause drastic seizure phenotypes. These differences are presumed to be due to differing levels of expression of other genes that modulate the excitation phenotype (Miller, A. R. et al., Genes, Brain, and Behavior (2014), Vol. 13, pp. 163-172; Mistry, A. M. et al., Neurobiology of Disease (2014), Vol. 65, pp. 1-11; and Hawkins, N. A. et al., Epilepsy Research (2016), Vol. 119, pp. 20-23).
In the brain, NaV1.2 and NaV1.6 are primarily expressed in excitatory glutamatergic neurons. Both channels are especially dense in the action initial segment (AIS), a region of the neuron adjacent to the neuronal soma that acts to integrate inputs and initiates action potential propagation to the soma and the distal dendrites (Royeck, M. et al., J. Neurophysiol. (2008), Vol. 100, pp. 2361-2380; Vega, A. V. et al., Neurosci. Lett. (2008), Vol. 442, pp. 69-73; and Hu, W. et al., Nat. Neurosci. (2009), Vol. 12, pp. 996-1002). NaV1.6 tends to be especially densely localized the early AIS (distal from the soma) where it is thought to act to trigger action potential initiation. NaV1.2 is more highly localized to the segment of the AIS most proximal to the soma. Mutations in both SCN2A (NaV1.2) and SCN8A (NaV1.6) have been linked to epilepsy and cognitive delay. The effects of the mutations are diverse both at the level of the impact on channel function, and on the patient phenotype. Both NaV1.2 and NaV1.6 are also expressed in peripheral neurons. NaV1.6 is especially dense at the nodes of Ranvier of myelinated neurons, where it is critical for maintaining salutatory conduction and high speed neuronal signaling.
Only a handful of NaV1.2 mutations have been described, but they are primarily linked with central nervous system pathologies, especially epilepsy (Kearney, J. A. et al., Neuroscience (2001), Vol. 102, pp. 307-317; Zerem, A. et al., European Journal of Paediatric Neurology: EJPN: Official Journal of the European Paediatric Neurology Society (2014), Vol. 18, pp. 567-571; Fukasawa, T. et al., Brain & Development (2015), Vol. 37, pp. 631-634; Howell, K. B. et al., Neurology (2015), Vol. 85, pp. 958-966; Saitoh, M. et al., Epilepsy Research (2015), Vol. 117, pp. 1-6; Samanta, D. et al., Acta Neurologica Belgica (2015), Vol. 115, pp. 773-776; Carroll, L. S. et al., Psychiatric Genetics (2016), Vol. 26, pp. 60-65; and Schwarz, N. et al., Journal of Neurology (2016), Vol. 263, pp. 334-343). The epilepsy mutations are presumed to be primarily gain of function mutations, meaning that they lead to an increase in the amount of sodium current and thereby increasing excitability. Establishing the impact on channel function in vivo beyond reasonable doubt is challenging and some of these mutations may yet lead to loss of function phenotypes.
Mutations in SCN8A have likewise been reported to show a range of gain and loss of function effects on the NaV1.6 channel though, for NaV1.6, most mutations examined have been associated with gain of function phenotypes. Mutations in NaV1.6 have been linked with epilepsy and autism spectrum disorders (Trudeau, M. M. et al., Journal of Medical Genetics (2006), Vol. 43, pp. 527-530; Veeramah, K. R. et al., Am. J. Hum. Genet. (2012), Vol. 90, pp. 502-510; Vaher, U. et al., Journal of Child Neurology (2013); de Kovel, C. G. et al., Epilepsy Research (2014); Estacion, M. et al., Neurobiology of Disease (2014), Vol. 69, pp. 117-123; Ohba, C. et al., Epilepsia (2014), Vol. 55, pp. 994-1000; Wagnon, J. L. et al., Human Molecular Genetics (2014); Kong, W. et al., Epilepsia (2015), Vol. 56, pp. 431-438; and Larsen, J. et al., Neurology (2015), Vol. 84, pp. 480-489). The best described SCN8A mutant patients have a syndrome known as early infantile epileptic encephalopathy, 13 (EIEE13). Over 100 EIEE13 patients have been identified. Patients typically present with intractable seizures between birth and 18 months of age. Patients have developmental and cognitive delay, and motor impairment often associated with chronic muscular hypotonia. The most severely impacted patients never gain sufficient motor control to walk. Many are not verbal. Less severe phenotypes learn to walk and talk but are motor-impaired and miss cognitive and social milestones. Most of the identified mutations are missense mutations, and it is assumed that the specific functional impact of the mutation contributes to the variability in the phenotype, though genetic background is also likely involved (Larsen, J. et al., Neurology (2015), Vol. 84, pp. 480-489). In contrast to SMEI patients, anecdotal evidence suggests that anti-epileptic drugs that target voltage-gated sodium channels non-selectively can ameliorate symptoms in EIEE13 patients, though no controlled clinical trials have been completed (Boerma, R. S. et al., Neurotherapeutics: The Journal of the American Society for Experimental NeuroTherapeutics (2016), Vol. 13, pp. 192-197). While phenytoin does seem to provide efficacy for EIEE13 patients, it does so at a cost. Efficacy is only achieved at very high doses where the significant adverse effects are tolerated only because the patients are in such dire need. Adverse effects commonly associated with phenytoin therapy include hepatic necrosis, hypertrichosis, nervousness, tremor of hands, numbness, dizziness, drowsiness, tremor, depression, confusion, fatigue, constipation, vertigo, ataxia, mental status changes, myasthenia, mood changes, restlessness, irritability, and excitement. It seems likely that a drug that selectively targets NaV1.6 would retain efficacy while reducing its adverse event burden.
Loss of function mutations in SCN8A in mice lead to a phenotype known as motor endplate disease (med) and multiple mutations and phenotypes were linked to the med gene region prior to the identification of the SCN8A gene (Burgess, D. L. et al., Nat. Genet. (1995), Vol. 10, pp. 461-465). Mice with SCN8Amed mutations have varying degrees of muscle hypotonia, consistent with the degree of dysfunction of the NaV1.6 function. Mice with the SCN8Amed/jo have NaV1.6 channels that have a loss of function, but not null, phenotype. SCN8Amed and SCN8Amed/jo mice are resistant to seizures induced by chemical insult (flurothyl, kainic acid, and picrotoxin) (Martin, M. S. et al., Human Molecular Genetics (2007), Vol. 16, pp. 2892-2899; Hawkins, N. A. et al., Neurobiology of Disease (2011), Vol. 41, pp. 655-660; and Makinson, C. D. et al., Neurobiology of Disease (2014), Vol. 68, pp. 16-25). Curiously, when SCN8Amed/jo mice are crossed with SCN1Anull mutant mice to produce a mouse that is heterozygous for both the SCN1Anull allele and the SCN8Amed/jo allele the double mutant mice have a much improved seizure and cognitive phenotype than those with only an SCN1Anull mutation (Martin, M. S. et al., Human Molecular Genetics (2007), Vol. 16, pp. 2892-2899). Such mice have a spontaneous seizure and death rate similar to wild type mice and their seizure threshold after chemical insult is also increased. A similar result occurs upon crossing mice with missense mutations of SCN1A (a model for GEFS+) and mice with SCN8A loss of function mutations. Having a single allele of SCN8Amed/jo protected the GEFS+ model mice from seizures and premature death (Hawkins, N. A. et al., Neurobiology of Disease (2011), Vol. 41, pp. 655-660). The ability of SCN8A knock down to improve seizure resistance is not limited to knockouts where the gene is globally absent throughout animal development. Knock down of SCN8A in adult mice either globally or specifically in the hippocampus via a CRE-LOX inducible knockout approach also improved resistance to electrically and chemically induced seizures Makinson, C. D. et al., Neurobiology of Disease (2014), Vol. 68, pp. 16-25). These data suggest that the suppression of inhibitory signaling caused by decreased NaV1.1 current can be offset, at least in part, by suppressing excitatory signaling via decreased in NaV1.6 current.
Voltage-gated sodium channel antagonism is the most common mechanism of widely prescribed antiepileptic drugs (AED's) (Ochoa, J. R. et al., Sodium Channel Blockers. In: Antiepileptic Drugs (2016), Vol. (Benbadis, S., ed) Medscape News & Perspectives). Carbamazepine, Eslicarbazepine, Oxcarbazepine, Lacosamide, Lamotrigine, Phenytoin, Rufinamide and Zonisamide are all believed to act primarily by blocking that function of NaV channels. Despite the presumed mechanism of action, these drugs are relatively promiscuous. They block all NaV channel isoforms indiscriminately, thus block of NaV1.1 would be expected to proconvulsant. Block of NaV1.6, and perhaps NaV1.2, would be anticonvulsant. In addition to sodium channels, these compounds also block other targets, including voltage-gated calcium channels. Selective NaV antagonists that spare NaV1.1 and other off-target receptors are expected to have both improved efficacy and therapeutic index relative to the currently available NaV blocking drugs.
There is therefore an unmet medical need to treat epilepsy and other NaV1.6 associated pathological states effectively and without adverse side effects due to the blocking of other sodium channels, such as NaV1.1 and/or NaV1.5. The present invention provides methods to meet these critical needs.